Outer Step 0. Inner Step 162. Batch Size 464

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5156190,
-    5156198,
-    5156203,
-    5156208,
-    5156213,
-    5156218,
-    5156222,
-    5156228,
-    5156233,
-    5156237
+    5156269,
+    5156273,
+    5156278,
+    5156282,
+    5156288,
+    5156293,
+    5156298,
+    5156303,
+    5156307,
+    5156313
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 154,
+  "inner_step": 162,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6de57a417037a20092a6f4fcd602ec8f59b1c493760bbd4fdcac54e4dbc6709
+oid sha256:e8cee6c626b3ce741e9e518da886349771412cd3b995a5f7610ce12fe21c293b
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a57292481ecdd1d6445f4d87b9f799ddb94af7191beb0e7f6216dbd62b78a8ee
+oid sha256:c43974b4d3ea93a7d6f914323f2830ccd2a11bdf2d88405465ff7449adca2281
 size 4040701744